Physical AI

SuperMap: Spatio-Temporal SLAM for Visual-Language Navigation

SuperMap is a living spatial memory for embodied AI that perceives the world, remembers its evolution, and supports reasoning and action, for visual-language navigation. Presented at RSS 2026.

Actuator Reality Shaping for Zero-Shot Sim-to-Real Robot Learning

Sim-to-real transfer in robot learning is often limited by discrepancies between the ideal actuator dynamics assumed during policy training and the nonlinear, hardware-dependent behavior of physical motors. While conventional approaches attempt to bridge this gap by increasing simulator fidelity through system identification, domain randomization, or learned actuator models, we introduce an alternative paradigm: actuator reality shaping. Instead of modifying the simulator to match the real world, our method shapes the closed-loop behavior of physical actuators to match the idealized second-order reference dynamics used in simulation. By equipping each joint with a two-degree-of-freedom feedforward–feedback controller, we decouple reference-response shaping from robust stabilization, thereby providing a standardized actuator interface for reinforcement learning policies. As a result, policies trained only with the prescribed reference model can be deployed zero-shot on real hardware without task-level fine-tuning or learned actuator models. We validate the approach on a single-joint high-gear-ratio servo under external loads and a 7-DOF robotic arm reaching task, where actuator reality shaping substantially reduces sim-to-real tracking error and improves zero-shot task performance compared with standard servo-control and representative real-to-sim-to-real baselines. We further demonstrate zero-shot transfer on a wheeled-legged robot driving over a slope and a humanoid robot walking, suggesting that actuator reality shaping can serve as a reusable interface for robot learning across diverse hardware platforms. Project page: https://syamamori.github.io/ActuatorRealityShaping.github.io/

Satoshi Yamamori, Koji Ishihara, Kenjiro Minamikawa

DL-SLAM: Enabling High-Fidelity Gaussian Splatting SLAM in Dynamic Environments based on Dual-Level Probability

Recent advances in 3D Gaussian Splatting (3DGS) have enabled significant progress in dense dynamic Simultaneous Localization And Mapping (SLAM). Prevailing methods typically discard predefined dynamic objects, ignoring that transiently static objects offer valuable geometric constraints for pose estimation. A recent work attempts to leverage this potential by employing per-pixel uncertainty maps to quantify the magnitude of motion. While this approach enables transiently static objects to enhance pose estimation, it erroneously integrates these objects into the static map, resulting in persistent artifacts. Moreover, its reliance on purely geometric information leads to ambiguous object boundaries in the uncertainty maps. To overcome these limitations, we present DL-SLAM, a monocular Gaussian Splatting SLAM system built upon a novel dual-level probabilistic framework. Our method computes dynamic probability maps by combining semantic and geometric information. These pixel-level probabilities are lifted to 3D and aggregated to derive an object-level dynamic probability for each instance. Object-level probability enables the categorical pruning of dynamic Gaussians, resulting in an artifact-free static map. The static map, in turn, provides a geometrically consistent guidance to refine the pixel-wise probabilities, enhancing their reliability. Experimental results demonstrate that DL-SLAM outperforms existing approaches, improving tracking accuracy by up to 13% while generating high-fidelity semantic maps.

Ziheng Xu, Qingfeng Li, Xuefeng Liu

NVIDIA SimFoundry turns real scenes into simulation worlds

NVIDIA GEAR SimFoundry builds simulation-ready worlds from a single image or video for real2sim2real robot training.

FADA: 2-Minute Humanoid Dynamics Adaptation

FADA separates planning from inverse dynamics, freezes the planner and fine-tunes only the IDM with about two minutes of target rollouts to adapt humanoid control to terrain, payload and actuator shifts.

Play2Perfect: Play Pretraining for Precise Assembly

Play2Perfect uses play pretraining to make precise assembly 33x more efficient.

VLK: Learning Humanoid Loco-Manipulation from Synthetic Interactions in Reconstructed Scenes

VLK synthesizes paired vision-language-kinematics supervision inside 3DGS-reconstructed real scenes: it generates navigation and object-interaction trajectories with privileged scene info, renders egocentric views after the fact, and produces 48,000 paired trajectories to train a policy predicting Unitree G1 whole-body motion, enabling sim-to-real perception-based humanoid loco-manipulation.

Yen-Jen Wang, Jiaman Li, Sirui Chen

GaussLite: Online Task-Conditioned 3D Gaussian Splatting for Real-Time Robotic Mapping

MIT presents GaussLite, the first 3DGS mapping system that takes a natural-language task and allocates representation capacity online by task relevance: a one-shot LLM parser extracts target and anchor objects, Grounding DINO + FastSAM ground them per frame into relevance masks that steer seeding density, gradient flow and initial scale. At matched Gaussian budget and real-time 4 Hz mapping, ROI PSNR improves by +2.72 dB on Replica and +2.23 dB on real campus scenes; multi-agent maps fuse via per-voxel voting on active-optimization counts, beating concatenation by +3.42 dB while sharing only 7.08% of the map.

Thomas, Annika, Peterson, Mason, How, Jonathan P.

FADA: Few-Shot Domain Adaptation via Dynamics Alignment for Humanoid Control

High-precision humanoid control is limited by target-domain dynamics mismatch, where the same control objective can induce different realized motions under changes in terrain, payload, or actuator response. Existing methods either pursue zero-shot transfer through domain randomization or in-context adaptation without target-domain specialization, or require heavy adaptation pipelines that leverage target-domain data, such as model calibration, residual learning, or policy retraining. In this paper, we present FADA (Few-Shot Domain Adaptation via Dynamics Alignment), a three-stage Planner-Inverse Dynamics Model (Planner-IDM) framework for few-shot adaptation in humanoid control. FADA first trains an oracle policy with privileged information and then distills the oracle behavior into a deployable Planner-IDM student through DAgger. At deployment, FADA freezes the planner and finetunes only the IDM using approximately 2 minutes of target-domain rollouts with standard supervised learning. Rather than requiring optimal demonstrations or rewards, FADA uses the paired actions and observations that are observed during these rollouts as supervision, aligning the IDM's action generation with target-domain dynamics. Experiments show that FADA outperforms both in-context and end-to-end adaptation baselines, improving task performance under dynamics shifts and enabling real humanoid robots to execute diverse high-precision whole-body tasks. Implementation details and qualitative hardware rollout videos are available at https://lecar-lab.github.io/FADA-humanoid/.

Angchen Xie, Nikhil Sobanbabu, Ishayu Shikhare

TopoRetarget enables zero-shot sim-to-real pen spinning on Wuji Hand

TopoRetarget builds a dexterous-manipulation RL pipeline from human demonstrations to real-robot execution. Combined with RL tracking, it achieves zero-shot sim-to-real pen spinning on the Wuji Hand.

TetherIA Aero Hand: Open-Source Robotic Hand Picks a Card via Teleoperation

TetherIA open-source Aero Hand robotic hand picks the top card from a deck and places it back via real-time teleoperation, showing fine grasping, controlled release and stable motion. Code and CAD are open-sourced.

Dexterous Hand Origami vs Japanese Master

Can dexterous robot hands fold paper airplanes better than a Nippon Origami Association master?